[RK-Net]Joint Representation Learning and Keypoint Detection for Cross-view Geo-localization

Prerequisites

Python 3.6
GPU Memory >= 8G
Numpy > 1.12.1
Pytorch 0.3+
scipy == 1.2.1
[Optional] apex (for float16) Requirements & Quick Start

Getting started

Installation

Install Pytorch from http://pytorch.org/
Install Torchvision from the source

git clone https://github.com/pytorch/vision
cd vision
python setup.py install

[Optinal] You may skip it. Install apex from the source

git clone https://github.com/NVIDIA/apex.git
cd apex
python setup.py install --cuda_ext --cpp_ext

Dataset & Preparation

Download University-1652 upon request and put them under the ./data/ folder. You may use the request template.

Demo

import torch
from model import USAM

input = torch.randn(128, 512, 16, 16)
usam = USAM()
output = usam(input)
print(output.shape)

Train & Evaluation

Train & Evaluation University-1652

python train.py --name RK-Net --share --extra --stride 1 --fp16; 
python test.py --name RK-Net

Default setting: Drone -> Satellite

Trained Model

You could download the trained model at GoogleDrive. After download, please put model folders under ./model/.

Citation

@article{lin2022,
  title={Joint Representation Learning and Keypoint Detection for Cross-view Geo-localization},
  author={Lin, Jinliang and Zheng, Zhedong and Zhong, Zhun and Luo, Zhiming and Li, Shaozi and Yang, Yi and Sebe, Nicu},
  journal={IEEE Transactions on Image Processing (TIP)},
  doi = {10.1109/TIP.2022.3175601},
  note={\mbox{doi}:\url{10.1109/TIP.2022.3175601}},
  year={2022},
  }
}

@article{zheng2020university,
  title={University-1652: A Multi-view Multi-source Benchmark for Drone-based Geo-localization},
  author={Zheng, Zhedong and Wei, Yunchao and Yang, Yi},
  journal={ACM Multimedia},
  year={2020}
}

Sj-Yuan / RK-Net